An Active Learning Algorithm Based on Existing Trainina Data

摘要

A multilayer perceptron is usually considered a passive learner that only receives given training data. However, if a multilayer perceptron actively gathers training data that resolve its uncertainty about a problem being learnt, sufficiently accurate classification is attained with fewer training data. Recently, such active learning has been receiving an increasing interest. In this paper, we propose a novel active learning strategy. The strategy attempts to produce only useful training data for multilayer per- ceptrons to achieve accurate classification, and avoids generating redundant training data. Furthermore, the strategy attempts to avoid generating temporarily useful training data that will be- come redundant in the future. As a result, the strategy can al- low multilayer perceptrons to achieve accurate classification with fewer training data. To demonstrate the performance of the strat- egy in comparison with other active learning strategies, we also propose an empirical active learning algorithm as an implemen- tation of the strategy, which does not require expensive compu- tations. Experimental results show that the proposed algorithm improves the classification accuracy of a multilayer perceptron with fower training data than that for a conventional random selection algorithm that constructs a training data set without explicit strategies. Moreover, the algorithm outperforms typi- cal active learning algorithms in the experiments. Those results show that the algorithm can construct an appropriate training data set at lower computational cost, because training data gen- eration is usually costly. Accordingly, the algorithm proves the effectiveness of the strategy through the experiments. We also discuss some drawbacks of the algorithm.